ollama_embed

Embed texts. LOW-LEVEL primitive. Returns raw 768-dim vectors for one text or a batch. WARNING: raw vectors can overflow MCP tool-output limits on large batches — for concept search prefer ollama_embed_search (ephemeral candidates) or ollama_corpus_search (persistent corpora); both return ranked hits, not raw geometry. Use this only for building external vector indexes (sqlite-vss, pgvector) where you need the vectors themselves. Envelope emits a warnings[] entry when the serialized payload crosses ~500KB.

Parameters

Parameter	Type	Required	Default	Description
input	string \	string[]	yes	—

Full input schema

The JSON Schema below is generated from the same zod schema the server validates against at the wire — it cannot drift from runtime behavior.

{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "type": "object",
  "properties": {
    "input": {
      "anyOf": [
        {
          "type": "string",
          "minLength": 1
        },
        {
          "minItems": 1,
          "maxItems": 256,
          "type": "array",
          "items": {
            "type": "string",
            "minLength": 1
          }
        }
      ],
      "description": "Text or batch of texts to embed."
    }
  },
  "required": [
    "input"
  ]
}

Envelope

Every tool returns the standard envelope — tier_used, model, tokens_in/tokens_out, elapsed_ms, backend provenance, warnings. See Envelope & tiers.

Source

Schema + handler: src/tools/embed.ts · registration: src/index.ts

---

Back to the Tool Reference.